HH models integration errors

[linusconrad]

Never mind found this old entry about wrong signs.

Dynamic current leads to runaway values

Hello, I’m trying to inject a current which is dependent on the membrane voltage. I’ve already created a TimedArray of conductance values and multiply this by Vm to get the current. However, during the run both current and voltage quickly become too large or NaN (depending on integration method). I suspect this is due to the current and voltage amplifying each other. Is there a way to update the current based on the voltage one time-step in to the past, or an even better way? Minimal code to r…

Checking and fixing those did the trick